How we describe coronavirus patients matters
This research inspired an award-winning poem, i Am A.I. and Thou.
Open data sets are vital for assessing the pandemic’s threat to local communities. Given the particularly decentralized response to the virus in the United States, individual, family, business, or institutional decisions drive the overall risk. While open science efforts compile authoritative data for medical professionals, we wondered what was available to the general public. Many non-medical professional may turn to local open data as sources of information.
New York University’s Digital Interests Lab investigated what coronavirus information governments made available to the general public as open data. Principal investigator, Dr. Anne L. Washington completed the work with NYU students Nicole Contaxis, Hope Muller, and Joshua Arrayales who have essays in this collection. Other researchers included Molly Nystrom and Dr. Rachel Kuo.
Who has COVID-19? This deceptively simple question was difficult to answer across jurisdictions in the open data sets available.
A central question for risk assesment during a pandemic is the rate of infection and the number fatalities. Some businesses and organizations will need to compare between jurisdictions. Many U.S. regionals hubs run cross state borders, such as the tri-state region around New York, New Jersey and Connecticut. In June 2020 we collected and analyzed open data sets related to COVID-19 patients in 50 states.
We found few consistencies in race, gender, and even in age reporting. We found outdated categories that excluded as much as organized. Respiratory diseases make older people particularly vulnerable so not being able to consistently track age is an immediate alarm. People from non-white categories are highly likely to work in essential businesses yet we have poor classification schemes for these identities. While there may be some impact on different hormonal and reproductive bodies, we are not tracking those characteristics well enough to differentiate risk.
The Digital Interests Lab research team members each explore a demographic characteristic to understand how the breadth of variation across our collection of open data sets. The first essay is a reflection on dual or non-binary identities with reference to the HIV/AIDS epidemic. The second essay tackles how age is treated across these data sets particularly in comparison to clasification standards that organize medical publications. The third essay explores which populations have no choice than “other” which renders their unique identity invisible.
Our findings point to open data variation across jurisdictions, however we identified several nationwide summer 2020 projects that reported demographic statistics: CDC Weekly Updates • Johns Hopkins University Racial Data Transparency • APM Research Lab and The Atlantic COVID Racial Data Tracker.
While we applaud efforts in national reporting, there is still a need to understand localized impacts for all populations.
Our preliminary evidence suggests that demographic categories need to be normalized for crisis reporting in future. Furthermore, outdated categories could compromise advanced data analysis. We argue for consistent data reporting standards and updated categorization of contemporary body identities.
These improvements may support the confidence of populations currently rendered invisible in the data. Together these efforts would improve the resuse of open data.
Categories of COVID Summer 2020 Research Team
Dr. Anne L. Washington — Principal Investigator
Dr. Rachel Kuo — Research Team Lead
Joshua Arrayales — Research Assistant
Nicole Contaxis — Data Librarian
Hope Muller — Data Scientist and Data Visualization
Molly Nystrom — Data Scientist
And thanks to Dr. Lauren Rhue and Dr. David Morar for additional feedback. The team Data Scientists are enrolled in NYU’s Master of Science in Applied Statistics for Social Science Research . This is the inaugural project from the Digital Interest Lab https://www.digitalinterests.org/.
I am grateful for the interest and support of the The Writer’s Fund at the Open Data Institute, London, UK. https://theodi.org/article/winners-odi-writers-fund-for-black-history-month-2020. and finally, for those of you reading this, wishing you and yours well in these difficult times.
— Anne L. Washington, http://annewashington.com